Natural language processing news

Equinor adds natural language processing to operational planning tool. Max Plank Institute open sources its Ambiverse natural language understanding software. Australian National Data Service releases Swagger OpenAPI-based vocabulary services. Open University and Springer Nature release Computer Science Ontology.

A posting on the Equinor (ex-Statoil) website describes how the company has added a natural language processing component (NLP) to OPT, its operational planning tool. NLP is used to locate hard to find information in text data such as incident reports. Context has been built into the NLP tool by training from dedicated, skilled users from Oseberg South to do digital highlighting. Localizing the NLP component to Norwegian meant building it practically ‘from the ground up’. The NLP component includes a knowledge graph of information in work orders or incident reports from several plants. Hyperlinks allows the users to open DNV GL Synergi cases directly from the tool. The NLP is work in progress and currently returns many ‘less than relevant’ results. But its developers believe it will mature to become a ‘great asset’ in the future. OPT runs in Equinor’s Azure cloud. Tessela and Norwegian Bouvet contributed to the Equinor NLP development.

The Max Plank Institute has released the Ambiverse Natural Language Understanding suite as open source software. AmbiverseNLU includes named entity recognition, entity and concept linking and knowledge graph components for several languages. A demo is available.

The Australian National Data Service, a partnership between Monash University, the Australian National University and the Commonwealth Scientific and Industrial Research Organisation has released a new vocabulary services API. The API enables programmable access to a number of controlled vocabularies for research. The API is built using the Swagger OpenAPI toolset for producing and consuming structured data.

The Knowledge Media Institute, a joint venture of The Open University and Springer Nature, has released the Computer Science Ontology (CSO), a ‘large-scale, automatically generated’ ontology of research areas in the field of computer science. The current version incorporates 14 thousand topics derived from a dataset of some 16 million articles. The CSO extends the BIBO ontology and is released under a Creative Commons license.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.